best_wflow_preds_vi: Get "best" parameter sets for each modeling approach (RF, DT,...

In rethinkpriorities/rp-r-package: Rethink Priority's R Package

Get "best" parameter sets for each modeling approach (RF, DT, GLM-EN), and also generating metrics of fit on training data

Description

Get "best" parameter sets for each modeling approach (RF, DT, GLM-EN), and also generating metrics of fit on training data

Usage

best_wflow_preds_vi(
  regression_workflow_results,
  outcome_var,
  classification = FALSE,
  metric = "rmse",
  train_sample = train,
  test_sample = test,
  vi = TRUE
)

Arguments

classification	is a logical to tell us if this is a classification model. in which case we want the ROC curve stuff
metric	is our measure of fit, I guess, like rmse or roc_auc
vi	if we want variable importance measures

Details

returns a tibble of the workflow ('output' is a 'fit workflow') with the best parameters, predictions, true outcomes, and some metrics

Examples

Used in best_wflow_preds_vi, also...
best_dt_preproc <- best_wflow_preds_vi(preproc_results, metric = 'roc_auc', classification = TRUE,
                                       outcome_var = "showup",
                                       train_sample = eas_all_train,
                                       test_sample = eas_all_test)

rethinkpriorities/rp-r-package documentation built on Jan. 7, 2023, 10:11 p.m.